Data storage subsystem

Abstract

A method for copying information from a first storage subsystem to a second storage subsystem is disclosed. The first and second storage subsystems are provided in a data storage system. The method comprises transmitting first data block from the first storage subsystem to the second storage subsystem, the first storage subsystem being associated with a first host computer and the second storage subsystem being associated with a second host computer; and transmitting first attribute information from the first storage subsystem to the second storage subsystem without intervention from the first host computer.

Description

CROSS-REFERENCES TO RELATED APPLICATIONS [0001] The present application is related to and claims priority from Japanese Patent Application No. 2002-163705, filed on Jun. 5, 2002. BACKGROUND OF THE INVENTION [0002] The present invention relates to a data storage system including a plurality of storage subsystem and a method of storing data therein. [0003] In recent years, the computer storage systems are configured to handle a huge amount of data. In addition, data is updated very frequently in these computer systems. A method for backing up large amounts of data and restoring a system to its normal state when an error occurs are important concerns in this technology field. One method used to address this concern has been a remote-copying technique. According to the technique, the data storage subsystems (external storage units), each having a magnetic disk array, are disposed away from each another. These data storage subsystems are then connected to each another through a communica...

AI Technical Summary

Benefits of technology

[0015] To make both of original and copied attribute information match completely as described above, the resynchronization must be achieved between both original and copied information, for example, after a remote-copy operation, stopped once, then restarted or after the copy direction is changed. To achieve the resynchronization, a difference of attribute information generated between those two operations is copied while the host computer does not know it in the embodiment of the present invention, thereby the attribute information is resynchronized quickly.

[0016] Recently, some users come to expect using of the above described functions for remote-copying over a plurality of data storage subsystems to improve the error resistance. The embodiment of the present invention also realizes the reinote copying.

[0021] According to the present embodiments, attribute information can be copied to a back-up data storage subsystem, so that the back-up side subsystem uses the attribute information automatically without intervention from the host computer. This eliminates complicated exchanges of attribute information between the host computers associated with the source subsystem and the destination subsystem.

[0022] Furthermore, according to the present embodiments, data and attribute information can be transferred to a destination data storage subsystem synchronously, i.e., in the order issued by the host computer, so that the data and the attribute information in the duplicate side subsystem can be updated with the received data and attribute information fast without the host computer involvement. If a difference occurs in any attribute information between stop of a remote-copy and restart of the remote-copy and / or between a forward remote-copy and its backward remote-copy, only the difference is copied. Without requiring the host computer involvement. Attribute information can thus be re-synchronized quickly between those stop and restart operations.

[0023] In addition, attribute information can be copied among a plurality of data storage subsystems, which makes it easier to remote-copy the attribute information among the plurality of data storage subsystems, thereby a system with higher error resistance properties can be formed more easily.

Problems solved by technology

The copy operation has thus been complicated.

This is why basically the updating order cannot be kept between attribute updating and data updating, although the updating order has been achieved with difficulty when the program in the subject host computer manages the order in close cooperation with the subject data storage subsystems.

However, the cooperation is not only difficult, but also comes to generate a large overhead.

Images